Effect Sizes

Approximated From Bushman and Anderson

Cohen’s Standards

Cohen (1992) proposed effect size standards for correlations of .10 (small), .30 (medium), and .50 (large).

Bushman and Anderson

In contrast, Bushman and Anderson (2001) ask us to think about the correlations of many associations that we think are serious or important from an intervention, public policy, or public health perspective.

Effect sizes are approximated from a Figure in Bushman and Anderson (2001).

Effect Sizes

library(readr)

effectsizes <- read_csv("effectsizes.csv")

library(plotly)

plot_ly(effectsizes,
        y = ~Association,
        x = ~Correlation,
        color = ~Correlation,
        text =  ~Correlation,
        name = "Effect Sizes",
        type = "bar")

References

Bushman, Brad, and Craig Anderson. 2001. “Media Violence and the American Public: Scientific Facts Versus Media Misinformation.” The American Psychologist 56 (June): 477–89. https://doi.org/10.1037//0003-066X.56.6-7.477.

Cohen, Jacob. 1992. “A Power Primer.” Psychological Bulletin 112 (July): 155–59. https://doi.org/10.1037/0033-2909.112.1.155.